"Frontmatter". In: Plant Genomics and Proteomics

S
EQUENCING
S
TRATEGIES AND
A
UTOMATION
The development of automated sequencing strategies and the reduction in
the cost of sequencing have allowed the design of experiments that rely
heavily on the acquisition of nucleic acid sequence information. This reduc-
tion in cost has also meant that redundant sequencing is no longer neces-
sarily a bar to identifying the sequence of interest. Therefore, shotgun
sequencing strategies that result in the same sequence being generated many
times to ensure a complete sequence assembly is now a viable option. The
complete sequence of a BAC can be obtained by the sequencing of subcloned
random fragments that have been isolated from sheared DNA. These
sequences are then assembled electronically to generate the linear sequence
of the BAC clone. Obviously, one potential drawback or problem in this
approach would be the presence of multiple copies of identical repeats
within the BAC, longer that the normal sequence read, that would be diffi-
cult to assemble. These gaps can be filled by using other strategies if neces-
sary. At least two approaches are available. In the first approach, regions of
the shotgun reads that are unique, and not close to the end of the read, can
be used to generate primers for additional sequencing reactions. This enables
the sequence to be read through repetitive regions provided they are short
enough to be covered in a single sequencing run. With improvements in tech-
nology this distance is now extending to more than a kilobase in length. The
second approach is to saturate the target sequence with a transposon and
use a primer close to the edge of this transposon to generate sequence reads.
This latter approach can get out of repetitive regions because the read can
start in the repetitive sequence and, provided the insertion is positioned suf-
ficiently close to the end of that region, read into the adjacent low-copy-
number sequence. Although the cost of generating the DNA sequence may
no longer be a limiting factor for almost any project, informatics challenges
associated with this rapid generation of sequence data are likely to provide
significant bottlenecks in data analysis for the foreseeable future.